issues
search
luigibonati
/
mlcolvar
A unified framework for machine learning collective variables for enhanced sampling simulations
MIT License
89
stars
24
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Improve transform tutorial coverage
#158
EnricoTrizio
closed
3 weeks ago
1
Doc minor fixes
#157
EnricoTrizio
closed
3 weeks ago
1
Fixed deltaF bug
#156
EnricoTrizio
closed
1 month ago
1
Update to latest micromamba
#155
EnricoTrizio
opened
1 month ago
0
Multiple dataset doc issue
#154
EnricoTrizio
closed
3 weeks ago
0
Numpy 2 support
#153
EnricoTrizio
opened
1 month ago
0
Numpy 2 update temporary fixes
#152
EnricoTrizio
closed
1 month ago
1
Committor updates
#151
EnricoTrizio
closed
1 month ago
1
Add utils for computing and plotting deltaGs
#150
EnricoTrizio
opened
3 months ago
0
Add link to latest doc to readme
#149
EnricoTrizio
closed
3 weeks ago
0
Missing transform tutorials
#148
EnricoTrizio
closed
3 weeks ago
0
Add SequentialTransform
#147
EnricoTrizio
closed
4 months ago
1
Test PyPI version via CI
#146
luigibonati
closed
4 months ago
2
Improvement committor module
#145
EnricoTrizio
closed
4 months ago
1
Bugfixes v1.1.1
#144
luigibonati
closed
4 months ago
2
The code is not compatible with torch 2.3, but is set to require the latest torch
#143
Alexander-the-Evil
closed
4 months ago
5
Register TDA target parameters as loss module buffers
#142
recisic
closed
4 months ago
1
Link to documentation in README
#141
recisic
closed
4 months ago
1
Multiple preprocessing transforms wrapped in sequential failing in training
#140
EnricoTrizio
closed
4 months ago
1
Initialize committor masses
#139
EnricoTrizio
closed
4 months ago
0
Set default activation of committor model back to tanh
#138
EnricoTrizio
closed
4 months ago
0
Committor doc
#137
EnricoTrizio
closed
4 months ago
0
Remove underscore in activation functions
#136
EnricoTrizio
opened
5 months ago
0
Make last layer activation of committor an explicit block
#135
EnricoTrizio
closed
5 months ago
3
linear activation function gives errors
#134
EnricoTrizio
closed
4 months ago
2
Bug fix transform and committor
#133
EnricoTrizio
closed
5 months ago
1
Add committor code
#132
EnricoTrizio
closed
5 months ago
1
Minor bugs fix May '24
#131
EnricoTrizio
closed
5 months ago
1
CI tests running on MacOS-13 and not MacOS-latest
#130
EnricoTrizio
opened
6 months ago
0
Fix issues in CI
#129
EnricoTrizio
closed
6 months ago
2
CI are failing to due to out of memory issue
#128
luigibonati
closed
6 months ago
2
I am having a set of features which includes RMSD, RMSF, SASA, (x,y,z) co-ordinates, RoG, Avg H-Bonds formed over the simulation in a seperate set of dataframes.
#127
vishnuthumpudi
opened
6 months ago
1
Can not compile some models to torchscripts with torch verion 2.
#126
jintuzhang
closed
1 month ago
2
I am using GROMACS as a molecular dynamics simulation tool and planning to use either MDAnalysis or mdtraj for analyzing collective variables. My question is whether these libraries support reading .xtc or .trr trajectory files generated from GROMACS simulations for identification of collective variables.
#125
vishnu-thumpudi2108
closed
6 months ago
1
Extension of transform module
#124
EnricoTrizio
closed
5 months ago
1
created a new class for SimulAELDA
#123
charlotte-chaps
closed
7 months ago
3
wrong PLUMED input example in tutorials
#122
EnricoTrizio
closed
5 months ago
1
Remove scipy version requirement from pyproject
#121
andrrizzi
closed
8 months ago
1
Interpretability module: sparse linear models via LASSO
#120
luigibonati
closed
4 months ago
2
Bug fix in nn.feedforward
#119
EnricoTrizio
closed
9 months ago
3
Dropout does not work
#118
geraseva
closed
9 months ago
1
Fix device issue checkpoint test
#117
EnricoTrizio
closed
9 months ago
1
Fix bug in lr scheduler configure
#116
EnricoTrizio
closed
9 months ago
1
Setting scheduler crashes
#115
andrrizzi
closed
9 months ago
0
Fix DeepLDA regularization
#114
luigibonati
closed
10 months ago
1
Fix #103: CVs cannot be loaded from checkpoint
#113
andrrizzi
closed
10 months ago
1
add sensitivity tutorial to doc
#112
luigibonati
closed
10 months ago
0
Add lr_scheduler options
#111
EnricoTrizio
closed
10 months ago
1
Fix bug unrepresented state.
#110
andrrizzi
closed
10 months ago
1
remove version check on scipy
#109
luigibonati
closed
10 months ago
1
Next